D
Deleted member 174368
Guest
They haven't even announced any new consoles yet, let alone the hardware specs. It's going to be awhile.
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
AMD Radeon HD 6870 & HD 6850 Video Card Review @ [H]
- Thread starter FrgMstr
- Start date
Exactly, I didn't think it was that much of an issue.
Lol, wow. Not only do you continually ignore the rebuttals I've made that show how wrong you are, you insist on showing again and again that you have absolutely no clue what you're talking about.
The reviewer chose to use MSI Afterburner to overclock and not CCC, and Afterburner does not yet support the 68xx cards, so they do not downclock to 2D idle but low power 3D when overclocked. That's reviewer's fault, not the cards. Again, if you knew a single thing about hardware, you would have recognized this.URL="http://www.overclockersclub.com/reviews/amd_6800series/7.htm"]http://www.overclockersclub.com/reviews/amd_6800series/7.htm[/URL]
Awww shucks, so when I'm not gaming, which is the majority of the time the computer is on, I'm actually using LESS power. Another reason to pick a 470.
So you're saying you think that a GTX 470, overlcocked 42%, only consumes an extra 20W over stock? What do you think is more probable, that the reviewer listed the wrong number or that this card defies the laws of physics? Answer this question in your next response so the forum can know whether or not to outright ignore you from now on.
For the couple hours of gaming it consumes 40Watts more. Wait, did we read that right? 40 Frickin Watts!! Yeah, so much for that "double the power consumption" even when overclocked with "heavily voltage modded" LOL theory...
Here's a hint for reference: http://www.legionhardware.com/articles_pages/palit_geforce_gtx_470,11.html a GTX 470 overclocked only to 760MHz core, nevermind 852MHz core, already consumes an extra 60W.
As the entire review showed, the majority of the time the GTX 470 @ 852MHz core is <10-15% ahead of the 6870. How many people do you think actually have their GTX 470's at 852MHz? How come you consider that but not the 6870 at 1200MHz that I linked? The truth hurts, doesn't it? You fell into NVIDIA's trap, bought an inferior card, and now you have to live with it. Don't think you're fooling anyone here, we're all well aware of what the situation is
.Though the HD4890 is certainly no slouch it is definitely starting to show its age in the light of these midrange cards. Man, they look good. I saw a HD6850 for $179 posted in another thread and that blew my mind. Bring on the HD6900's! Man, it was a long wait to the 22nd, now we gotta do it all over again.
PooprScooper
Weaksauce
- Joined
- Aug 24, 2010
- Messages
- 66
Exactly, I didn't think it was that much of an issue.
Lol, wow. Not only do you continually ignore the rebuttals I've made that show how wrong you are, you insist on showing again and again that you have absolutely no clue what you're talking about.
The reviewer chose to use MSI Afterburner to overclock and not CCC, and Afterburner does not yet support the 68xx cards, so they do not downclock to 2D idle but low power 3D when overclocked. That's reviewer's fault, not the cards. Again, if you knew a single thing about hardware, you would have recognized this.
So you're saying you think that a GTX 470, overlcocked 42%, only consumes an extra 20W over stock? What do you think is more probable, that the reviewer listed the wrong number or that this card defies the laws of physics? Answer this question in your next response so the forum can know whether or not to outright ignore you from now on.
Here's a hint for reference: http://www.legionhardware.com/articles_pages/palit_geforce_gtx_470,11.html a GTX 470 overclocked only to 760MHz core, nevermind 852MHz core, already consumes an extra 60W.
As the entire review showed, the majority of the time the GTX 470 @ 852MHz core is <10-15% ahead of the 6870. How many people do you think actually have their GTX 470's at 852MHz? How come you consider that but not the 6870 at 1200MHz that I linked? The truth hurts, doesn't it? You fell into NVIDIA's trap, bought an inferior card, and now you have to live with it. Don't think you're fooling anyone here, we're all well aware of what the situation is
So, instead of accepting the truth, you try to make up a story about how the review site that YOU linked doesn't know how to review anymore. Maybe you should of read the article that you claimed supported your ideas before posting it. You're just sounding pathetic trying to make up any story to support your claims at this point.
How many people OC their 470s you ask? You're asking this on a OCing site. So the answer is, just about everyone. You seem to think that ONLY AMD graphics cards have the ability to idle. The 470 is about 25 Watts in idle and 210W when maxed out. Yeah, so much for the doubling the power of a 6870 idea...
You seem to be under the assumption that just because the 480 is a power guzzler that the 470 is too. You've made it clear to everyone that you have no idea what you're talking about and are just speaking irrationally.
Not only did I save $19 but apparently having more FPS is "inferior" to you. Just keep telling yourself that fanboy. I buy what's good at the time of purchase, not which brand I blindly support. My 5870 has served its purpose and we'll see whether a 6950 is worth switching back for or not once the 580 comes out.
Lord_Exodia
Supreme [H]ardness
- Joined
- Sep 29, 2005
- Messages
- 7,008
Arguments aside, I have sort of a question for you, why go from 5870 to GTX 470. Kind of a side or down grade hmm. Was there something the 5870 wouldn't do for you that the 470 will or does? I've never read someone who went from a single 5870 to a 470. Dual yes, single no. Just want to see what your motivation was.
There is no need for anyone to be retarded. It's the sign of a small-minded person who, instead of just enjoying his own purchase, has to jump down others' throats for their own purchase decisions to make himself feel superior.
If you like the 6870 more than the 470, good for you. But who cares? At the pricepoint that the 470 very quickly fell to, it's been a good deal for a long time. The additional RAM is also a nice assurance at triple monitor resolutions.
If you like the 6870 more than the 470, good for you. But who cares? At the pricepoint that the 470 very quickly fell to, it's been a good deal for a long time. The additional RAM is also a nice assurance at triple monitor resolutions.
eh? Who's got the 6870 in hand already? I don't think anyone here has one yet, may be wrong tho.
But sure see a few justifying their 470's on an AMD subforum which is quite funny actually. Came back here after 24hrs and it's still the same folks rehashing the same stuffs.
The article was referenced for performance comparisons and showed quite well how a GTX 470 is at best competitive performance-wise with a 6870 while being inferior in all other metrics. The fact that this rips the rug out from under your arguments, once again, and your sequential temper tantrum, shows that you have nothing, and that this discussion is above you. If you would like to actually prove any of your points, go right ahead. So far your best arguments have been "I found this number!" all the while with a clear indication that you have no idea what the number means, never mind how it relates to an actual video card's function. Sorry, laughing like a child and typing in all caps doesn't prove your point, no matter how much you would like it.
I asked how many people run their GTX 470's at 852MHz core, don't change the question.
Where did I say that? Go quote it.
Are you talking about stock performance or overclocked performance? You continually mix up the two, and seem to believe that overclocking doesn't change power consumption, which is ridiculous. Power consumption increases approximately linearly with clockspeed and quadratically with voltage, welcome to EE 101. If you want to argue stock performance and power consumption, do that, or argue overclocked performance and power consumption, but you can't mix and match and to create impossible, and ridiculous, situations. At stock, the 6870 consumes about 50W less than a GTX470 and is on average 3% slower across most gaming resolutions, with the 6870 being faster at higher resolutions: http://www.techpowerup.com/reviews/HIS/Radeon_HD_6870/29.html .
You haven't made a thing clear. I've disproved every one of your ill-contrived arguments and you have not even attempted to rebut any of them. Ad hominem attacks and shouting without any proof just further detract from your arguments. That crap may work on other forums, but at [H] people pay attention and know what they're talking about.
It's a good thing NVIDIA got you hook, line, and sinker. You enjoy your 1% more performance and all the heat, noise, and inferior image quality that comes with it.
The 470 does triple monitors?
PooprScooper
Weaksauce
- Joined
- Aug 24, 2010
- Messages
- 66
Because of the price drop to $220. It's cheaper to sell the 5870 and get 2 470s than a second 5870 while still having almost as good framerates. I also wanted to try BC2 in 32xCSAA while taking only a minimal performance hit in fps. I'm using two 460s in another machine for CUDA and I was thinking of buying a separate 9800gtx just for PhysX with the 5870 but now I'll be able to do it all with the 470. I'm not so concerned with the extra $1-2/mo in electricity like some other posters apparently and the computer isn't in a room that's 5x5 feet where the so called "extra heat" won't even be noticeable.
Brent_Justice
Moderator
- Joined
- Apr 17, 2000
- Messages
- 17,755
If anyone is curious:
I don't put much into that, was just curious about possible tessellation performance improvements. 6870 is higher than 5850, with less streaming processors, so looks like some real improvements there that might translate to games.
Code:
Unigine Heaven Benchmark 2.1
Default Config
Normal Tessellation - 2560x1600 - 4X AF - High - DX11
6870:
FPS: 23.7
Scores: 597
Min FPS: 16.2
Max FPS: 47.5
gtx 460 1gb
FPS: 20.8
Scores: 523
Min FPS: 14.4
Max FPS: 40.0
5850:
FPS: 20.6
Scores: 518
Min FPS: 13
Max FPS: 43.6
6850:
FPS: 20
Scores: 505
Min FPS: 14.3
Max FPS: 39.6I don't put much into that, was just curious about possible tessellation performance improvements. 6870 is higher than 5850, with less streaming processors, so looks like some real improvements there that might translate to games.
Not trying to pick on you here Pooper but ive read the arguements and it really does look like Mr. K6 has got you beat. If you like Nvidia then thats cool, im actually a bit saddened really by Nvidia's fall over the past few generations because i've been an Nvidia consumer since the TNT days.
It probably doesnt take a 5x5 room either for those cards to heat a room. I've got a pair of 5870s and after a good gaming session the room is noticeably warmer then the rest of the home and this is a 12x12 room. Kyle also gives some good information here ---> http://hardocp.com/article/2010/03/26/nvidia_fermi_gtx_470_480_sli_review/7
Last graph on the page is most relevant to my post in that, under load all these cards get hot, but the GTX4x0 cards DO get hotter and while it may only be a few degrees, its quite a noticeable amount, especially in a home.
If you buy 2 then yes, it will do triple monitor gaming. Nvidia's marketing calls it NVSurround.
Last edited:
Actually, power consumption is more important than core temp. If both cards are at 80C, but one uses 150 W, and the other one 250 W, the latter one will heat the room more.
fullvietFX
[H]ard|Gawd
- Joined
- Sep 1, 2004
- Messages
- 2,015
So does the 6870 = 5770 replacement?
Yes it does. and likewise, 6850=5750 replacement
necrophile
[H]ard|Gawd
- Joined
- Sep 6, 2008
- Messages
- 1,243
Which reviews are you reading?
Uhhh....Say what?
http://www.techpowerup.com/reviews/HIS/Radeon_HD_6870/29.html
But, if you consider the 57xx series mainstream, then yeah, you're right. Yeah, the 68xx are the new mainstream AMD cards.
Last edited:
Good point, though I likened it in my mind to a fan blowing the output temps shown in the graphs, electrical usage levels would be more of a root cause.
You just don't get it
. The 470 is a high end card which was retailing for $400 earlier this year and is now forced mid range thanks to AMD's superiority. It is a low yield card and very expensive to manufacture. I have not seen the card near $220 anywhere which kind of makes your argument moot as well. When one company's mid range effort can compete with anothers high end, I think the word ass kicking is pretty accurate. As for the power savings, even at a conservative $2 per month savings, you would be looking at saving $72 over a 3 year lifespan which is about average for most consumers. The devil is in the details.
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
it was $220 at Newegg today and sold out.You just don't get it
As for the power savings, even at a conservative $2 per month savings, you would be looking at saving $72 over a 3 year lifespan which is about average for most consumers. The devil is in the details.
PooprScooper
Weaksauce
- Joined
- Aug 24, 2010
- Messages
- 66
How long have you been buying video cards? Today's highend is tomorrow's low end. A 8800GTX was a high end model as well, until the newer models came out and knocked it down. The 470 has been out for awhile and it's bound to be bumped down as newer models are about to be released(580). A 6870 isn't THAT far behind a 5870 which had been a high end model for awhile now. To be honest, a 6850 looks more like a mid-range card anyways. The 6870 is more lower-highend. Try reading page 1 for the link to the 470 priced @ $220. You can buy SLI combos from newegg at around $240 as well.You just don't get it
As for the power savings, even at a conservative $2 per month savings, you would be looking at saving $72 over a 3 year lifespan which is about average for most consumers. The devil is in the details.
Of course all video cards put out heat, but I have this fancy thing called A/C which keeps the house cool. The difference between 2 5870s and 2 470s is minimal and air is circulated throughout the house anyways.
AMD classified the 6850 and 6870 as mid-range cards. I think it was stated on one of slides they were spewing out. The high-end cards are the 6950, 6970 and 6990. The 6950 being the low high-end model.
BTW, I reckon the 580 will costs around $500+.
BTW, I reckon the 580 will costs around $500+.
SylarPowers
[H]ard|Gawd
- Joined
- Apr 9, 2009
- Messages
- 1,263
I keep reading mixed opinions in various reviews about the stock 6870 fan noise. Anyone get to hear theirs yet? Some say its quiet, some say its as loud as a 5850-5870 (My XFX 5850 was fine for general use to me, but I never got the time to game on it enough to make it hot), while others say its loud. Even though I bought a XFX 6870 (which I needed as I had no card at all), I will likely get the 6970 Vapor X Sapphire model down the line because of the techs proven improvements in the fan noise / temp areas. I think people confuse normal operation and likely gaming sound output with unlikely worse case scenarios. With the efficiency of these cards, I think any full speed fan noise worse case scenario is unlikely.
My FT02 manual says it prefers external exhaust due to the proximity of the intakes, something Anandtech reinforced before - they say in general it helps to have some distance for non external exhaust or the two fans fight each other somewhat in some way. But the Vapor X stuff gets massive drops in temps usually, so I think that would make up for any fan turbulence caused by internal exhaust. As in I think a Vapor X 6870 or 6970 would work fine in a FT02, just somewhat less effectively. Not to mention I'm rocking three intake AP-181 Fans in combo with 1850rmp GTs on the CPU and Exhaust areas, which seemingly help a ton. And I don't see the AP-181 fans being confused / impeded by the open style GPU fans at all.
Just my thoughts after reading up on my XFX 6870 purchase.
My FT02 manual says it prefers external exhaust due to the proximity of the intakes, something Anandtech reinforced before - they say in general it helps to have some distance for non external exhaust or the two fans fight each other somewhat in some way. But the Vapor X stuff gets massive drops in temps usually, so I think that would make up for any fan turbulence caused by internal exhaust. As in I think a Vapor X 6870 or 6970 would work fine in a FT02, just somewhat less effectively. Not to mention I'm rocking three intake AP-181 Fans in combo with 1850rmp GTs on the CPU and Exhaust areas, which seemingly help a ton. And I don't see the AP-181 fans being confused / impeded by the open style GPU fans at all.
Just my thoughts after reading up on my XFX 6870 purchase.
Last edited:
Battleneter2
Gawd
- Joined
- Nov 3, 2006
- Messages
- 973
Yea actually my fac last gen O/C GTX 275 can play Just Cause 2 Predator just fine 40+ FPS, sure I have to turn down AA, but I can see much of a diff above 4x AA at these resolutions anyway. Metro is the exception but the game sucks. Sure not exactly a old DX9 vard only, but the other guy is right, games are now behind the hardware.
Still trying to convince myself I need a upgrade, but the latest game I am playing is Fallout Vegas and that sits on like 80FPS on Ultra settings, so that's not helping the upgrade itch much.
"Almost" pressed the butting on a 6870, but just......cant...... do ..... it
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
you do realize that the gtx275 is a pretty high end last gen DX10 card don't you? he said an old DX9 card. and Metro 2033 played just fine on high DX10 AAA settings at 1920x1080 with my gtx260 so it should be no prob with a gtx275.
why on earth would you even consider a 6870 since that would only be about a 30% improvement? with your setup the upcoming 6900 cards will be a much better choice.
SylarPowers
[H]ard|Gawd
- Joined
- Apr 9, 2009
- Messages
- 1,263
Any guesses on how long it will take Sapphire to make a Vapor X 6870 or Vapor X 6970?
PooprScooper
Weaksauce
- Joined
- Aug 24, 2010
- Messages
- 66
Yea, I was only calling a 6870 lower-highend because at the moment, the 6900s aren't out and it's barely behind a 5870. The 6850 is definitely much more mid-range powered compared to the higher end models that are available today.
$500 for a 580 sounds about right. Nvidia/Intel has always been price gougers.
You're right in that todays hi end is tomorrow's low/mid end. A company's newer cards will routinely bump down their older cards - that's pretty much a given.
However, in the case of the 470, it didn't get bumped down by any of NVidia's cards since the successor isn't out yet, it got bumped down by a competitor's card. I think that's what Alienate was refering to. Trust me, if AMD did not release the 6850/6870 or waited even a month before releasing them, 470 prices would have stayed the same.
In other words Nvidia's hand was forced by AMD.
To add further when the Fermi's came out the "todays hi end is tomorrows mid/end" didn't quite apply to the 5870. How come? There was hope that it would bring alot to the table and would force AMD to lower prices on 5850 and 5870 just like what we're seeing with the 460/470 - .that didn't happen. 5870 prices barely budged because it didn't need to. What does that say about Fermi's impact on the hi end marketplace?
Last edited:
PooprScooper
Weaksauce
- Joined
- Aug 24, 2010
- Messages
- 66
Right. What's wrong with them lowering prices though? Happens all the time between the two companies.
Fermi brought the amazing 460s that overclocked like no other before. Two of them in SLI were cheaper and performed better than a 5870/480. And the 480 is still king of the single GPU market. It's probably never going to be cheaper than a 5870 so ATI doesn't need to reduce their price. It sure is a gas guzzler though.
This is killin me. I have a 5870 now. Was hoping that after xmas the price would come down, and I could get another for xfire. It looks like it will be like my 280gtx in another rig. It just morphed into a 285, and now you can hardly find them, AND they still cost over $200.
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
5870 crossfire on a 3.8 dual core and at 1680x1050? there will be many games where you would not even notice that other 5870 with your cpu. heck at that res you could even lose performance in some cases just due to your cpu having to handle the overhead. crossfiring cards of that level need a quad preferably an i5/i7 if you actually want to push them fully especially in more cpu intensive games. even now you are not fully pushing a single 5870 in very cpu intensive games at just 1680.
I do not think that crossfire overhead on a dual cpu clocked to 3.8 will exceed the benefit of a second video card (a 5870). If you have any data or links to prove this, please provide them. The wildest claim of all is a loss of performance (lower fps's) going from a single card to cross fire.
Here's some links to testing, rather than just giving you my opinion.
http://alienbabeltech.com/main/?p=19601&all=1
I liked the conclusion on this one:
It’s reasonable to expect anyone with a single high-end GPU like a GTX470 to be running something like 1920×1200 with 2xAA, while CF/SLI owners will obviously go higher, like 2560×1600 with 4xAA. After all, that’s the whole point of buying high-end graphics cards. Nobody buys high-end graphics cards to run games at 1280×1024 with no AF or AA so they can show four cores getting 200 FPS while two cores “only” get 150 FPS.
If you’re currently on a decent dual-core platform with a low or middle class video card, absolutely do not be afraid of upgrading to a high-end graphics card if you have a 1920 (or better) monitor and/or you like using AA. When games are configured to use the highest playable settings, in the vast majority of cases the graphics card will influence gaming performance the most, often to the point of completely bottlenecking the system.
Of course if you’re building a brand new rig from scratch, get yourself a mid-range quad-core (e.g. an i5 750) so you can future-proof yourself for 2-3 years. In addition, it goes without saying that you should buy the fastest graphics card you can afford.
http://alienbabeltech.com/main/?p=4090
http://www.overclock.net/pc-games/659536-contagion-review-dual-core-vs-quad.html
http://www.legionhardware.com/articles_pages/metro_2033_performance_guide,6.html
.
Last edited:
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
its not a a wild claim at all. if you know anything about sli/crossfire then you know this is true and can happen. the weaker the cpu and the lower the res the more likely it will happen. there have been reviews that have shown this in the past. I cant easily find some so for now have a look at this. even with a Core i7 980x at 4.0 and using fairly large resolutions 4 way and 3 way sli can be slower than 2 way sli due to cpu overhead. http://www.tomshardware.com/reviews/quad-sli-nvidia-surround-geforce-gtx-480,2745-7.html
"The problem is consistency, as it appears that CPU overhead has reduced the configuration’s maximum frame rate under moderate graphics load in a way that really hurts any analysis of average performance"
now take his much slower cpu and much lower res he could most certainly not gain a thing in some games. and in more cpu intensive games he could lose a few fps. heck with my E8500 at 3.8 I can lower my res from 1920 to 1280 and still get the same performance in GTA 4, Dragon Age, Bad Company 2 and some other games. now that's with just a single gtx260 so imagine running 4x the gpu power with 5870 crossfire and see how much of a bottleneck those cards would receive in cpu intensive games.
and here is come scaling for different cpus using 5870 crossfire. too bad they did not use most games but you can clearly see the i5/i7 getting way better performance compared to an Core 2 duo. sometimes nearly twice the frame rate even. http://www.legionhardware.com/artic...ssfire_cpu_scaling_performance_part_2,21.html
"For the most part the Core i7 9xx and Core i5 7xx series were in a league of their own. Something that we did also find surprising was the fact that for the most part the Core i3 5xx series kept pace with the Phenom II X4 on a clock-for-clock basis."
EDIT: Metro 2033 is one of the very few games where he would be able to really use that second 5870 at lower res even with his cpu.
just turning of cores on an i5/i7 doesn't tell you how it will run on another architecture. here is a good comparison where you see an i7 loses nothing really with 2 cores disabled in Bad Company 2 yet the Core 2 quad architecture does. http://www.legionhardware.com/articles_pages/gaming_the_core_debate,1.html
also they still had this to say..."So even a powerful processor such as the Core i7 920 still struggles in these games without the correct number of cores."
Last edited:
cannondale06, your first link contains no cpu dual vs quad testing, and the second link has to do w/ performance increase realative to clock speed.............that is exactly my point. Right now, speed, getting things >3.0, is more important than dual vs quad "in gaming" (not benchmark tests), both for single and dual gpu.
And, the 3.8 is definitely high even by today's standards. So, any differences we are talking about are on the order, of what 5%, 15%, or 25%. Certainly not "NO" gains, and certainly not Negative gains (losses).
Please give me a link that shows that a dual-core C2D, clocked at 3.8, would not see benefits of 25% - 45% in FPS's using a 5850 or a 5870 in crossfire.
Most of the problems w/ the 5000 series in Crossfire have been w/ the drivers/design, not bottle necking, if you are overclocked nicely.
There will always be a bottleneck somewhere in your system. So unless you have the budget to build new systems every 6 months you may as well lay out a plan for upgrade. What's important for you to replace first and go from there.
.
And, the 3.8 is definitely high even by today's standards. So, any differences we are talking about are on the order, of what 5%, 15%, or 25%. Certainly not "NO" gains, and certainly not Negative gains (losses).
Please give me a link that shows that a dual-core C2D, clocked at 3.8, would not see benefits of 25% - 45% in FPS's using a 5850 or a 5870 in crossfire.
Most of the problems w/ the 5000 series in Crossfire have been w/ the drivers/design, not bottle necking, if you are overclocked nicely.
There will always be a bottleneck somewhere in your system. So unless you have the budget to build new systems every 6 months you may as well lay out a plan for upgrade. What's important for you to replace first and go from there.
.
Last edited:
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
my first link was because you said it was a wild claim that adding another card could actually ever lower performance. it can.
you need to learn to read better because that second link compares many different cpus at different speeds. it clearly shows the i7/i5 to be superior. I have even put the conclusion up for you.
READ the results. for example even at a 2560 the i5 at 3.6 is nearly 75% faster than the E8xx Core 2 at 3.6. http://www.legionhardware.com/artic...ssfire_cpu_scaling_performance_part_2,18.html
there is no point in arguing if you cant look at the info right in front of you and understand it.
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
and remember I actually have an E8500 at 3.8. I have used a gtx470 in this system before and in cpu intensive games I picked up little to no performance over the gtx260 even at 1920. Far Cry 2, Bad Company 2(forced to DX10 for comparison) and GTA 4 to name a few did not move 1 single fps. some other games that did improve still had the same minimum framerate. of course some games like Just Cause 2, Clear Sky and Metro 2033 did improve though. now he is talking about the equivalent have having slightly more 2 of those gtx470 card with the same cpu so there will be some serious bottleneck in many games especially cpu intensive ones.
here you can see Dragon Age is more than twice as fast on an i7 at 3.5 than it is with an E8400 oced to 3.6. and that's just with a single 5870 so adding another 5870 would not really do anything since the cpu is already the limitation .http://www.pcgameshardware.com/aid,...rks-75-percent-boost-for-quad-cores/Practice/
when I first replied to him I thought he was at 1680 but he now says he is going to a 27 inch so that is likely 1920. at 1920 that would help quite a bit but the cpu would still limit the crap out of a second 5870 in very cpu intensive games since its not even coming close to fully using a single 5870.
here you can see Dragon Age is more than twice as fast on an i7 at 3.5 than it is with an E8400 oced to 3.6. and that's just with a single 5870 so adding another 5870 would not really do anything since the cpu is already the limitation .http://www.pcgameshardware.com/aid,...rks-75-percent-boost-for-quad-cores/Practice/
when I first replied to him I thought he was at 1680 but he now says he is going to a 27 inch so that is likely 1920. at 1920 that would help quite a bit but the cpu would still limit the crap out of a second 5870 in very cpu intensive games since its not even coming close to fully using a single 5870.
so is that going to be 1920x1200, 1920x1080 or 2560x1400? either way even at 1920 its not as bad as I was saying for 1680. I would still sell the 5870 and go with the fastest single card I could get though.
Last edited:
I never claimed that the "i" processor "family" was not faster than the C2D processor family. That's just too obvious to comment on further.
I responded to your claim that if this dude moved from a 5870 to a 5870 CF, that he might not see any improvement in FPs, or actually worse FPs (I'm paraphrasing). You did say that right?
I understand what you're saying, it was just your broad brush strokes that set me off...that's all.
I have no idea about your Nvidia upgrade experience....too many variables. But your experience is not the "common" experience. Most people that perform significant upgrades in their video card experience an improvement in fps, all things being equal.
It's a difficult discussion to have on a forum...I trust we are both Geeks and know a fair amount, right? We are both reading [H], right? And we both have our experiences, and we have researched different sources.
And as I said, there are always bottle necks, and trade offs to be made. You can create your own bottle neck by:
- change any hardware (ram, motherboard, hard drive, cpu, gpu, monitor)
- play a different game, or change it's settings (AA, etc.)
- change your overclock
- etc.,
There are other bottle necks besides the most obvious, cpu and gpu, like the monitor (limited to 60 fps, usually), and your eye. Also, many people choose use Vsync, which limits your fps.
Ahhhhh, the list goes on.
Good discussion, let's just give each other some respect, and perhaps chalk this up to minor differences in the width of our brush's....lol.
Edit: I agree w/ the single card approach..that's why my gaming machine has an 5870 and a Q9550. I have not upgraded yet, and may wait another full cycle of both the cpu and gpu.
.
I responded to your claim that if this dude moved from a 5870 to a 5870 CF, that he might not see any improvement in FPs, or actually worse FPs (I'm paraphrasing). You did say that right?
I understand what you're saying, it was just your broad brush strokes that set me off...that's all.
I have no idea about your Nvidia upgrade experience....too many variables. But your experience is not the "common" experience. Most people that perform significant upgrades in their video card experience an improvement in fps, all things being equal.
It's a difficult discussion to have on a forum...I trust we are both Geeks and know a fair amount, right? We are both reading [H], right? And we both have our experiences, and we have researched different sources.
And as I said, there are always bottle necks, and trade offs to be made. You can create your own bottle neck by:
- change any hardware (ram, motherboard, hard drive, cpu, gpu, monitor)
- play a different game, or change it's settings (AA, etc.)
- change your overclock
- etc.,
There are other bottle necks besides the most obvious, cpu and gpu, like the monitor (limited to 60 fps, usually), and your eye. Also, many people choose use Vsync, which limits your fps.
Ahhhhh, the list goes on.
Good discussion, let's just give each other some respect, and perhaps chalk this up to minor differences in the width of our brush's....lol.
Edit: I agree w/ the single card approach..that's why my gaming machine has an 5870 and a Q9550. I have not upgraded yet, and may wait another full cycle of both the cpu and gpu.
.
Last edited:
cannondale06
[H]F Junkie
- Joined
- Nov 27, 2007
- Messages
- 16,180
it is really going to depend on the game and settings of course so we could go back and forth all day. lol
lets just leave it alone since we both know there are pros and cons to it.
lets just leave it alone since we both know there are pros and cons to it.